• It’s not just the soft drinks

    A number of you took exception to my saying that banning large soft drinks wouldn’t work. Some of you think it’s not “prohibition”. Some of you think it’s still worth a try. Some of you absolutely think it’s going to work. I still disagree. I think it’s cherry-picking, it’s mostly for optics, and it’s not going to lead to a slippery slope of success.

    Here’s one reason why: a new study entitled, “Beverage patterns among Canadian children and relationship to overweight and obesity“:

    Sweetened beverage intake has risen in past decades, along with a rise in prevalence of overweight and obesity among children. Our objective was to examine the relationship between beverage intake patterns and overweight and obesity among Canadian children. Beverage intake patterns were identified by cluster analysis of data from the cross-sectional Canadian Community Health Survey 2.2. Intake data were obtained from a single 24-hour recall, height and weight were measured, and sociodemographic data were obtained via interview. Data on children and adolescents aged 2–18 years who met inclusion criteria (n = 10 038) were grouped into the following categories: 2–5 years (male and female), 6–11 years (female), 6–11 years (male), 12–18 years (female), and 12–18 years (male). χ2 test was used to compare rates of overweight and obesity across clusters. Logistic regression was used to determine the association between overweight and obesity and beverage intake patterns, adjusting for potential confounders.

    In short, this study was looking for a relationship between beverage consumption and obesity after adjusting for other factors. What did they find? First, here’s the unadjusted analysis. You’ll see age and sex categories in the left column. The other columns are the beverage cluster the children fell into, or the dominant beverage they consumed. For each age/sex group you get the percent of kids who are normal weight, overweight, and obese for each beverage.

    See a pattern? No? Me neither. For 6-11 yo females, being in the “soft drink” cluster gave you the lowest chance of being obese. Same for 12-18 yo males. But it’s all over the place. All of those p values on the right hand side show that in none of the age/sex groups was BMI classification associated with beverage cluster.

    Here are the results of the logistic regression. For the record, they controlled for age, total energy intake, ethnicity, a measure of sedentary activity, and sociodemographic characteristics:

    What you’re looking at here are odds ratios, and then 95% confidence intervals in parentheses after them. An odds ratio above 1 means that you’re more likely to be overweight and obese; and odds ratio less than 1 means you’re less likely. If the 95% confidence interval includes 1, then the results are not significant.

    Only one result was significant. Soft drinks were found to be significantly associated with a higher odds of overweight and obesity in boys 6-11 yo. No other relationships were seen for any beverages (including soft drinks) and any of the other age/sex groups. Looking at the prevalences in the first table I showed you that’s not surprising. But I find it fascinating that this relationship is totally missing in the 12-18 yo males. This means that either it’s a totally new phenomenon in young boys, who will continue to be obese, or this relationship disappears as they grow older.

    Regardless, it’s hard to look at these data and say that it’s definitively soft drinks that need to go. Why single them out? As I said on Wednesday, we need to fidn a way to get people to consume less calories overall, not from one beverage (halfheartedly) only in certain settings.


    Comments closed
    • I’m one of the earlier responders. I fit into these categories: this is not prohibition; it will not work; it’s worth a try because it could lead to a slippery (albeit very very bumpy) road to success.

      Like I said, this won’t work, and I agree that soft drinks alone won’t cut it. But we know quite well that they’re empty calories, so it’s likely that any comprehensive weight loss intervention will involve people cutting down on soft drinks (the full calorie versions).

      Also my read of the abstract (haven’t read the full paper, no access) is that their groups are people whose main beverage consumption is soft drinks, fruit juices, milk and “moderate”. What it shows is that people who mainly drink soft drinks aren’t disproportionately overweight. What is does not show is what happens if you make moderate reductions in soft drinks for all overweight individuals. So if I’m right, we could also say you haven’t necessarily proved your point either.

      There is in fact evidence that regular consumption of full calorie sweetened beverages (so I guess includes fruit juices) does increase the risk of overweight for preschoolers:


      So it may be that action on soda alone will have some positive effect. This is not prohibition enough for me to feel that this sort of ban is harmful – again, Singapore has low obesity rates but to get there they march people off to fitness class and/or the armed forces. That’s draconian. Bloomberg is just the unpleasant overbearing petty tyrant at the office by comparison.

    • You said that (re:significance for 6-11 yr old boys) that this is a new phenomen or they outgrow it.
      Another possibility is that they ran numerous test and got one significant result by chance. This explains it better than I could in this space: http://xkcd.com/882/

      • That is a possibility. I assumed they ran a correction fro multiple comparisons, but I admit I didn’t look closely!

    • Here’s the thing. I’m 44 years old. When I was a little kid Cokes in restaurants came in 6 oz glass bottles. Very cold. Very delicious. And when it was gone, it was gone.
      I’m trying to enforce the same “restaurant only treat” with soda with my kids — and they are served a minimum of 12 oz, usually 20 and you’ve got to beat the refillers back with a stick if you don’t want your kid to consume a gallon of soda at a meal. (And my kids are in that 6-11 boy group; the one who loves soda outweighs his identical twin by 15 pounds).
      It’s ridiculous. 32 oz sodas are ridiculous.
      I think the ban is worth doing to start recalibrating our eyes and our palates back to reasonable, healthy portion sizes. I think the slippery slope could lead us back to sanity.
      Remember, the super-sizing of all this stuff was done without any study of its health effects, without any research other than what the market would bear.

    • Despite this study (maybe Canadian kids just haven’t caught up with the American phenomenon yet), I still think soft drinks are near the top of the source of obesity list . And, I think Bloomberg’s initiative has already worked… a little. Disney changed their policy on advertisers.

      Making the super-sized drinks uncool is how you reduce their effect, and now merchants have to consider whether or not they’re doing something the community regards as anti-social when they promote those products.

      Ending the promotion is probably good enough for diminishing in a significant way this part of the problem. They may never need to enforce Bloomberg’s rule to get the desired effect. This state-led badgering is worth it if they don’t go crazy with enforcement.

      I recently bought an iced tea at a fast food joint; unsweetened, $1, came in only one size. Before it was half gone I could not hold the thing in one hand. They’re serving drinks in small buckets!

    • A large study from the US has shown sugar sweetened beverage consumption to be modestly associated with CHD (http://circ.ahajournals.org/content/early/2012/03/09/CIRCULATIONAHA.111.067017.abstract) and Type 2 Diabetes (http://www.ncbi.nlm.nih.gov/pubmed/21430119).

      I’m no expert on this literature, but I think that the policy discussions should mostly consider epidemiologic studies that concentrate on health outcomes most related to mortality and healthcare costs. A Relative Risk of 1.20 is relatively small and the studies could be prone to bias especially since it’s observational, but I’d prefer the debate to be on this literature rather than ones that concentrated on an intermediate endpoint that is prone to measurement error (BMI).

      It would be great to hear thoughts on the modeling studies that project increased mortality and healthcare costs due to sugar sweetened beverage consumption and whether this is sufficient evidence for a tax (or whatever):

      “Through the model, the researchers estimate the additional disease caused by the drinks has increased coronary heart disease healthcare costs by 300-550 million U.S. dollars between 2000-2010….Over the last decade, at least 6,000 excess deaths from any cause and 21,000 life-years lost can be attributed to the increase in sugar-sweetened drinks.”